Trading Statistics 2026: Real Data on Trader Performance

The commonly cited "90% of traders lose money" comes from a handful of academic studies across different markets, eras, and trader populations — each with a different methodology and headline number. The reality is more nuanced: profitability rates depend heavily on sample definition (active journalers vs everyone who ever opened an account), time horizon (30 days vs 12 months), and market (day traders fail at different rates than swing traders). In our sample of 8,400+ active TSB journal users with 90+ days of recorded trades, 31% are net profitable — higher than academic studies of all retail day traders, because active journaling self-selects for more disciplined traders.

This guide compiles the real trading statistics across three data layers: our internal anonymized dataset (what we can measure directly), peer-reviewed academic studies (what the research literature reports), and industry-reported data (broker and prop firm disclosures). Plus the survivorship-bias trap that inflates every statistic you see quoted, the win-rate vs reward-risk relationship that separates winners from losers, and the journaling frequency effect that's the most controllable variable in the dataset.

Internal statistics sourced from anonymized TSB journal data (8,400+ users, 90+ days active, 1.2M+ trade records, Q1 2026). Academic citations drawn from peer-reviewed research: Barber, Lee, Liu & Odean (2014) on Taiwanese day traders, Chague, De-Losso & Giovannetti (2020) on Brazilian equity day traders, Barber & Odean (2000) on US retail investors. External data sources: DALBAR Quantitative Analysis of Investor Behavior, CFTC retail forex trader disclosures. See methodology section for full sample definitions, limitations, and biases.

What Percentage of Traders Are Actually Profitable?

The headline number varies dramatically by sample definition. Comparing our dataset against the main academic studies:

Our Dataset (TSB Active Journalers, 90+ Days)

Comparison to Published Academic Studies

Why the Numbers Diverge

Four factors create the gap between "3% are profitable" and "31% are profitable":

• Sample definition. Academic studies typically include every account that ever traded, including people who tried for two weeks and quit. Active-journaler samples exclude quick quitters, inflating apparent success rates.
• Time horizon. Single-quarter measurements show higher profitability rates than multi-year measurements because short-term luck hasn't normalized. The Brazilian study's 3% figure comes from requiring 300+ trading days — most short-term "profitable" traders give back gains before hitting that bar.
• Cost accounting. Gross profitability can be 2-3x net profitability once commissions, spread, swap, and slippage are counted. The Brazilian 3% figure is net-of-costs and may be the most realistic for comparing across markets.
• Market structure. Day trading has different failure rates than swing/position trading, which has different rates than investing. Aggregating across timeframes produces meaningless averages.

The most intellectually honest summary: 3-15% of all people who seriously try day trading end up net profitable across multi-year windows, but that rate climbs substantially for traders who actively journal, review losses, and trade longer timeframes. Our 31% figure for active journalers is consistent with this interpretation — self-selection into serious practice materially improves outcomes, but the majority still lose money.

Win Rate Distribution Across Traders

Average win rate across all TSB users: 47%. But averages hide the distribution and the relationship to profitability.

Win Rate Buckets and Profit Factor

Why Win Rate Alone Is Misleading

The 40-50% bucket is the largest group but has the sub-1.0 average profit factor — meaning the median trader at these win rates loses money net. Traders in the 60%+ bucket have higher win rates but lower R:R, so profitability is similar to the 50-60% bucket. This is consistent with the academic literature: Barber & Odean (2000) found that overtrading and costs dominate win-rate differences in explaining why most retail investors underperform the index.

Reward-to-Risk Ratios: Where Profitable Traders Separate

Average R:R across all TSB traders: 1.3:1. For profitable traders specifically: 1.8:1. The gap is where most losing traders concentrate.

The R:R / Win Rate Breakeven Matrix

The 0.1 R:R Gap That Determines Most Outcomes

Traders in the 40% win rate bucket need 1.5:1 R:R to break even before commissions, but average only 1.4:1 actual. That 0.1 R:R gap — 100 basis points of reward efficiency — is the difference between slowly making money and slowly losing it. It sounds trivially small because the absolute number is small, but at 40% win rate it's the margin between being in the 31% profitable group or the 58% unprofitable group. Tighter stop management or slightly extended profit targets can close this gap without requiring better entries.

Does Journaling Actually Help Performance?

The frequency-performance correlation in our data is strong, though causation is harder to prove definitively:

Correlation vs Causation

Two interpretations both have support in the data:

• Selection effect: More disciplined traders are both more likely to journal daily and more likely to be profitable. Journaling is a symptom of discipline, not a cause of improvement.
• Causal effect: Journaling itself creates feedback loops that accelerate learning. Traders who force themselves to document decisions catch repeated mistakes faster.

The improvement-over-time metric (rightmost column) is harder to explain with selection alone: daily journalers improve their profit factor by 18% over 6 months while rare journalers improve by only 3%. If selection were the only driver, improvement rates would be similar once grouped by starting skill. The 18% vs 3% gap suggests at least some causal component to the journaling effect — consistent with the deliberate-practice literature showing that structured reflection accelerates expert skill development across domains.

Performance by Asset Class

TSB users trade multiple markets. Performance breakdown:

Why Asset Class Matters

Stock/options traders show the highest profitability rate (35%) in our sample, likely because equities trend more reliably over longer timeframes and options provide defined risk on the downside. Crypto traders have the highest average R:R (1.5:1) because of larger price swings, but the lowest win rate (44%) and profitability rate (27%) — volatility cuts both ways, and the 24/7 market creates overtrading exposure that weekend-closed markets don't.

Futures traders hit the middle on profitability (29%), which is consistent with the dominance of prop firm-funded futures traders in the category — prop firm rules constrain drawdown and position sizing in ways that mechanically keep some losing traders out of the sample.

The Hidden Deal-Breaker: The Survivorship Bias in Trading Statistics

Most trading statistics — including the ones on this page — suffer from survivorship bias. The accounts that blew up and disappeared aren't in the sample; the traders who quit in month 2 of a 6-month study aren't in the sample; the traders who never started journaling in the first place aren't in the sample.

How the Bias Inflates Every Number

• Account churn is invisible. Brokers report profitable-account percentages for currently-active accounts. Accounts closed after loss aren't in the denominator. This structurally overstates "% profitable" in any broker-sourced statistic.
• Quitter exclusion is the biggest bias. Studies requiring 90+ days of data automatically exclude traders who quit before 90 days — which is where the worst losses typically happen. A sample requiring 6+ months excludes even more of the bottom tail.
• Self-selection into journaling is severe. TSB users actively choose to journal. This is a ~10-20% fraction of all retail traders. Our 31% profitability figure is not generalizable to all retail traders — it's generalizable to active journalers.
• Platform selection adds another layer. Academic studies of Brazilian or Taiwanese traders sample specific brokers with specific customer profiles. US retail forex traders under CFTC rules are a different population again.

The Honest Number

Combining all available data with bias adjustments, the honest estimate: 5-15% of people who attempt active trading for 12+ months are net profitable after costs. The 31% figure for active journalers in our dataset is not a correction to that number — it's a statement that journaling self-selects for traders who are 2-3x more likely to be in the profitable tail. The base rate for the population who never start journaling is substantially worse than 31%, likely in the 5-10% range based on academic studies of full populations.

Practical read: If you're new to trading, don't benchmark against the 31% figure — that applies to traders already doing the work. Benchmark against the 5-15% academic range. If you're already journaling 4+ times per week, you've self-selected into the 31% bracket, and the relevant comparison is whether you're in the profitable half of that bracket.

3 Mistakes Traders Make Interpreting Statistics

Mistake 1: Comparing Your Win Rate Without R:R Context

Traders see "average win rate is 47%" and feel bad if their win rate is 42%. But the 42% trader with 2:1 R:R is more profitable than the 50% trader with 1:1 R:R. Win rate is not a standalone performance metric — it only makes sense in combination with R:R and cost drag. Before comparing your win rate to any benchmark, check whether your R:R is enough to support that win rate on a breakeven basis. The breakeven matrix above is the right comparison frame.

Mistake 2: Assuming "90% of Traders Lose" Means You Have a 10% Chance

The statistic describes outcomes across a full population, most of whom quit in month 1-3 without meaningful practice. It does not say that you, specifically, have a 10% chance of profitability if you do the work. Traders who journal 4+ times per week, review losses, and trade for 12+ months see profitability rates of 30-40% in our sample. The 10% figure describes the base rate for the population that doesn't do those things. Your effort level matters more than the population statistic.

Mistake 3: Cherry-Picking Statistics to Confirm Existing Beliefs

Trader A will quote "3% of traders are profitable" to justify quitting. Trader B will quote "38% of daily journalers are profitable" to justify continuing without changing anything. Both statistics are real; both are being misused. The honest frame: profitability rates are distributed along an effort/discipline axis. Wherever you currently sit on that axis, the relevant statistic is the conditional rate for your effort bracket, not any single headline number.

Who Should Skip Benchmarking Against These Stats

Benchmarking against aggregate statistics isn't the right framing for every trader. Specific profiles should measure differently:

• Traders under 90 days of live data. Short-horizon performance is dominated by variance, not signal. A 60% win rate over 30 trades means little; over 300+ trades it starts to reflect underlying skill. Don't compare yourself to any benchmark until you have 90+ days and 200+ trades recorded.
• Swing or position traders with low trade frequency. Our sample skews toward day traders (70%+ of trades). Swing traders with 5-10 trades per month have different baseline statistics (higher win rates, longer cycles) and benchmarks from active day-trading populations don't apply.
• Traders in a strategy transition. If you changed strategy within the last 90 days, your current statistics reflect a blended system that doesn't exist anymore. Benchmark only after 90+ days on a stable approach.
• Traders optimizing for variance-adjusted metrics. Sharpe ratio, Sortino ratio, and Kelly criterion require different baselines than win rate and profit factor. Our sample doesn't publish volatility-adjusted stats at the aggregate level.
• Prop-firm evaluation candidates specifically. Prop firm pass rates and funded trader statistics (see pass rate data) are a different population with different rule constraints. Retail trader statistics overstate achievable profitability at prop firms because of drawdown and consistency rules.

Methodology, Sample, and Limitations

For our full editorial and data process, see our editorial methodology. For the academic literature on retail trader profitability, the Barber & Odean papers cited above are the starting canon.

Final Verdict: The Honest Picture

The "90% of traders lose" statistic is directionally correct but imprecise. Across the available academic and industry data, 5-15% of people who seriously attempt active trading over 12+ months end up net profitable after costs. The rate climbs to 30-40% for traders who actively journal, review losses, and maintain disciplined practice — but the majority of this uplift comes from self-selection (disciplined people are more likely to journal and more likely to profit), not purely from journaling as a magic bullet.

The win rate / R:R relationship is the tightest signal in the data. Traders at 40% win rate averaging 1.4:1 R:R are slightly unprofitable; the same win rate with 1.6:1 R:R would be profitable. Closing the 0.1-0.2 R:R gap is the highest-leverage improvement available to most losing traders, typically via tighter stop management or extended profit targets — not via changing entries.

Three principles from this data:

• Base rate depends on effort bracket. Quoting a headline profitability number without stating the sample is misleading. "31% of active journalers" and "3% of full retail population" are both true for different populations.
• Win rate without R:R is meaningless. Use the breakeven matrix to contextualize your win rate before concluding you're over or underperforming.
• Survivorship bias inflates every quoted statistic. The traders not in the sample are typically the worst performers. Adjust accordingly.

For related analysis: why traders lose money for the failure-mode breakdown, performance analysis guide for calculating your own metrics, emotional trading patterns for the psychology side, revenge trading cost analysis for the single most expensive behavioral pattern, and prop firm pass rates for the funded-trader subset of this population.